Diabetes Duration and Subclinical Myocardial Injury: The Atherosclerosis Risk in Communities Study (ARIC)

Abstract

Background: Diabetes exerts adverse effects on the heart, and a longer diabetes duration is associated with greater heart failure risk. We studied diabetes duration and subclinical myocardial injury, as reflected by high-sensitivity cardiac troponin (hs-cTnT).

Methods: We analyzed 9052 participants without heart failure or coronary heart disease (mean age 63 years, 58% female, 21% Black, 15% with diabetes) at The Atherosclerosis Risk in Communities Study (ARIC) Visit 4 (1996 to 1998). Diabetes duration was calculated based on diabetes status at Visits 1 (1987 to 1989) through 4, or using self-reported age of diabetes diagnosis prior to Visit 1. We used multinomial logistic regression to determine the association of diabetes duration with increased (≥14 ng/L) or detectable (≥6 ng/L) Visit 4 hs-cTnT, relative to undetectable hs-cTnT, adjusted for demographics and cardiovascular risk factors.

Results: The prevalence of increased Visit 4 hs-cTnT was higher in persons with longer diabetes duration, from 12% for those with diabetes 0 to <5 years up to 31% among those with diabetes for ≥15 years (P for trend <0.0001). New onset diabetes at Visit 4 was associated with 1.92× higher relative risk (95% CI, 1.27-2.91) of increased hs-cTnT than no diabetes. Longer diabetes duration was associated with greater myocardial injury, with duration ≥15 years associated with 9.29× higher risk (95% CI, 5.65-15.29) for increased hs-cTnT and 2.07× (95% CI, 1.24-3.16) for detectable hs-cTnT, compared to no diabetes.

Conclusions: Longer diabetes duration is strongly associated with subclinical myocardial injury. Interventional studies are needed to assess whether the prevention and delay of diabetes onset can mitigate early myocardial damage.

Fig. 1.

Prevalence of increased high sensitivity troponin T across categories of diabetes duration.

Fig. 2.

Relative risk ratios (95% CIs) for increased high sensitivity troponin T by categories of diabetes duration. Regression model adjusted for age, sex, race-center, smoking, alcohol, body mass index, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, systolic blood pressure, triglycerides and estimated glomerular filtration rate.

Fig. 3.

Restricted cubic spline for the continuous association of diabetes duration with adjusted relative risk ratio for increased hs-cTnT (≥14 ng/L). Adjusted for age, sex, race-center, smoking, alcohol, body mass index, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, systolic blood pressure, triglycerides and estimated glomerular filtration rate. Diabetes duration in years modeled using a restricted cubic spline, with knots located at 0, 3, 9, and 18 years, and a reference group of no diabetes at ARIC Visit 4. A histogram of diabetes duration is also presented. New onset diabetes was associated with a RRR of 1.92 (95% CI, 1.27–2.91) for increased hs-cTnT.